Going beyond the A-constant for IOL calculations

Issue: October 2018

ByUday Devgan, MD

The idea seems reasonable: introduce a variable into lens power calculations in order to fine-tune the results and improve accuracy. The A-constant, which is a bit of a misnomer because it is a variable and not actually a constant, can help provide a way to adjust based on the IOL type — including refractive index and lens geometry — IOL position, calibration of biometry devices and even surgeon technique. Once this is honed for a particular surgeon and IOL, it can be used to adjust all IOL calculations for that surgeon so that, as a whole, all of the eyes undergoing cataract surgery can achieve a more myopic or more hyperopic outcome.

Adjusting the A-constant will change the IOL power with a 1:1 ratio, so that increasing the A-constant by 0.5 D means that the calculated IOL power will be increased by 0.5 D across the board for all eyes. For a particular model of IOL, many factors will affect the A-constant. If the lens geometry is such that the haptics have a posterior angulation, then the optic will sit deeper within the eye, and the A-constant will be higher than for an IOL that sits more anterior. IOLs are labeled for their optical power, which is measured outside of the eye, and the variability of how it sits within the eye, the refractive index and the lens design (plano-convex vs. bi-convex, for example) will change the effective refractive power in the eye. This is why two different IOL models will have different A-constants and thus different IOL calculations even though both of them measure the same optical power outside of the eye. If you know that a particular model of IOL has an A-constant of 119.2 but then you switch to a different IOL with an A-constant of 118.7, you must then also drop the IOL power by 0.5, which is the difference between 119.2 and 118.7 in order to maintain the same refractive outcome.

Many of the IOL-related variables will be the same even among different surgeons. This is why online databases such as the ULIB website (User Group for Laser Interference Biometry) calculate A-constants for many of the popular IOLs. This is a good starting point, but then it must be tailored to the individual surgeon. For changes in machine calibration, the premise is simple: If a surgeon has a keratometer that tends to measure lower than the actual corneal power, the A-constant can be increased to account for that. The same can apply for surgical techniques, such as the effect of capsulorrhexis morphology on the effective lens position of the IOL.

Figure 1. If we plot out our range of IOL calculations with regards to different axial lengths, we get a curve, shown in navy blue. If we decide to increase our A-constant to give a slightly more myopic postop result, we get the exact same curve but just raised higher, shown in orange. Decreasing the A-constant would simply lower the curve.

Figure 2. A better way to increase the accuracy of our calculations is to optimize the calculations for different sizes of eyes. This results in the curve shown in green, which has a different shape compared with the original navy blue curve. This shows us that longer axial lengths must be treated differently than shorter axial lengths.

Figure 3. We agree that axial lengths over 25 mm should have an adjustment made to the IOL power to increase accuracy. This pioneering work by Li Wang, MD, PhD, and colleagues shows that the adjustment at 25 mm is less than at 30 mm, with a variable adjustment depending on axial length. The same idea applies to all other eyes, including those with an axial length of less than 25 mm. This idea of adjusting for different types of eyes applies to all variables including axial length, keratometry, anterior chamber depth and more.

But there is a problem with applying the same A-constant to all IOL powers. Remember that changing the A-constant will affect all eyes and all IOL calculations equally, whether the eye has a short or long axial length. Increasing the A-constant simply lifts the exact same IOL calculation graph by the same amount across the board (Figure 1). We cannot treat all eyes the same. Long eyes are different from short eyes; to increase the accuracy of IOL calculations, we have to treat them differently. We can optimize our calculations for specific axial lengths, but doing so is far more complicated than just raising or lowering the A-constant (Figure 2). We can also take into account other variables beyond axial lengths, such as keratometry and anterior chamber depth by graphing in three dimensions. This more complicated math can be done with computer algorithms. The website www.IOLcalc.com offers this service for free using the Ladas Super Formula, and it can optimize based on a surgeon’s own data or, for unusual eyes, based on the large library data set in order to pool the data from many surgeons. In the past year, you have done cataract surgery for many patients with a 24 mm axial length, but how many have you done on eyes with an 18 mm axial length? By combining data from hundreds or even thousands of surgeons, we can amass a large database of even these unusual eyes.

In 2011, Li Wang, MD, PhD, and colleagues published a pioneering algorithm for adjustment of the IOL powers for highly myopic eyes. They chose to adjust the axial length of the eye, but they could have just as easily chosen to adjust the A-constant. This is a variable adjustment that has little effect at 25 mm and then progressively more adjustment as the axial length increases to 30 mm and beyond. But why does this idea of adjusting for different axial lengths stop at 25 mm? It does not, and the eyes that have axial lengths of less than 25 mm still need some form of variable adjustment to increase IOL calculation accuracy. And do not stop at just the axial lengths because we can also do adjustments for keratometry and anterior chamber depths (Figure 3).

PAGE BREAK

Recent work by David Cooke, MD, showed a comparison of nine different IOL calculation formulas wherein he sought to bring the mean error to zero by adjusting the A-constant. This worked for many formulas but not for the Ladas Super Formula because it goes beyond the A-constant and does not optimize all eyes the same way. Most IOL calculation formulas simply adjust the A-constant and shift the curve up or down (Figure 1), whereas the Ladas Super Formula actually hones the shape of the entire curve by optimizing eyes individually.

The A-constant was a reasonable starting point, but we are at the point where it is simply not a sufficient way to accurately hone our IOL calculations. We have to treat different eyes differently, and that means optimizing across the spectrum of axial lengths, keratometry, anterior chamber depths and more. The best way to show you this is with your own patients, calculations on www.IOLcalc.com and accurate results.

References:
Cooke DL, et al. J Cataract Refract Surg. 2016;doi:10.1016/j.jcrs.2016.06.029.
Holladay JT. J Cataract Refract Surg. 1997;doi:10.1016/S0886-3350(97)80115-0.
Wang L, et al. J Cataract Refract Surg. 2011;doi:10.1016/j.jcrs.2011.05.042.

For more information:
Uday Devgan, MD, is in private practice at Devgan Eye Surgery, Chief of Ophthalmology at Olive View UCLA Medical Center and Clinical Professor of Ophthalmology at the Jules Stein Eye Institute, UCLA School of Medicine. He can be reached at 11600 Wilshire Blvd. #200, Los Angeles, CA 90025; email: devgan@gmail.com; website: www.DevganEye.com.

Disclosure: Devgan reports he is a principal in Advanced Euclidean Solutions, which owns the Ladas Super Formula, Ladas Super Surface and www.IOLcalc.com.

Read more

Going beyond the A-constant for IOL calculations

Related Content